Network website traffic strangling making use of tc/tbf - unanticipated outcomes

In order to facilitiate some efficiency screening I am trying to strangle the ip website traffic on a certain equipment - I desire restriction it to 1Mbit/sec.

I have a different examination equipment where this command functions simply great:

tc qdisc add dev eth0 origin tbf price 1mbit ruptured 5kb latency 70ms peakrate 1.001 mbit minburst 1540

On the actual box where I intend to use this, I get quirkiness. The specific very same command restrictions traffic to 6Kbit/sec. Any kind of suggestions why?

unanme - a programs

Linux IPERF1 2.6.18-128.el5PAE #1 SMP Wed Dec 17 12:02:33 EST 2008 i686 i686 i386 GNU/Linux

We have a 3 various means of checking the throughput : The real application we are examining which reports the throughput we see. We additionally some Tivoli surveillance on a button and also we have an HTTP customer which downloads a documents from an HTTP web server on package concerned. All 3 approaches show 6Kbs when the TBF is used, and also several Mbs when we remove the TBF.

2019-05-13 05:10:52
Source Share
Answers: 1

Timer resolution could affect your outcomes, see here as an example. What takes place if you increase ruptured to 10kb and also remove peakrate and also minburst?

2019-05-21 05:44:04